Heart rate variability, long pitched as an early Alzheimer's clue, mostly failed to predict outcomes over 13 years

A Dutch cohort study of 269 cognitively unimpaired adults tested whether resting heart rate variability measured years earlier predicted later cognition, blood biomarkers or brain scans. It found almost no such link once coronary artery calcification was taken into account.

Resting heart rate variability (HRV) measured eight and thirteen years earlier was not associated with later cognitive test scores, most Alzheimer's disease blood biomarkers, or brain imaging measures in a cohort of 269 cognitively unimpaired adults, according to a study published in the Journal of Alzheimer's Disease. The one exception ran in the opposite direction researchers expected: among participants with high coronary artery calcification, higher HRV measured years earlier was linked to a lower serum Ab42/Ab40 ratio at follow-up, a pattern usually associated with more amyloid pathology, not less.

HRV, the beat-to-beat variation in heart rhythm that reflects autonomic nervous system activity, has drawn interest for years as a low-cost, wearable-friendly signal of the autonomic dysfunction seen in some dementias. Other cohorts have reported links between low HRV and faster cognitive decline; a 2024 analysis of 2,702 adults in the UK Whitehall II study, for instance, found low HRV associated with accelerated decline in global cognitive function over a decade. The new study was designed to test that idea more rigorously by also accounting for coronary artery calcification, a marker of cardiovascular disease that can independently affect both heart rhythm and brain health.

Researchers from the University of Groningen and colleagues drew participants from the Dutch Lifelines cohort and its Imalife and Memolife sub-studies. HRV, calculated from the root mean square of successive heartbeat interval differences, was measured at two early time points, 2007-2013 and 2014-2017, then tested against cognition, blood-based Alzheimer's markers, and MRI brain measures collected in 2022-2023, with participants split into groups with absent (score 0) or high (score 300 or above) coronary artery calcification on CT scans. Across the full sample, HRV predicted none of the cognitive, biomarker or imaging outcomes; the single significant association, in the high-calcification subgroup, had a modest effect size (beta -0.30, 95 percent confidence interval -0.47 to -0.13) and ran counter to the researchers' hypothesis.

The authors write that their findings argue for factoring in coronary calcification when HRV is studied as a marker of cognitive decline, and note the result should temper broader claims about HRV as an early Alzheimer's warning sign. The study does not test HRV as captured by consumer wearables, which use different sensors and shorter recording windows than the clinical electrocardiograms used here, and its 269 participants are a fraction of the size of some cohorts, such as Whitehall II, that have reported positive associations.

Researchers involved in the work say the result should inform how HRV is used in future dementia studies, not how any individual reads their own wearable data. Whether HRV holds any value as an Alzheimer's signal, they conclude, may depend heavily on which cardiovascular factors are measured alongside it.